Data Mining With WebFOCUS RStat

In this section:

Until recently, data mining was a branch of Business Intelligence (BI) used only by expert statisticians. Few people understood their mathematical methods. Thus, the statistical results were communicated only to upper management. But, increasingly, operational users are required to make decisions and take actions based on their expectations about the future.

Most reporting applications do a good job of recording what has happened. But that is just a rear-facing view of the business. They do not provide guidance about future actions. To compete effectively in the business world today, decision makers at every level of an organization need access to predictive modeling applications. Police officers need to determine where crimes are likely to occur so patrol cars can be in areas where they are most needed. Marketing managers need to predict who is most likely to respond to an email blast or ad campaign. Auto insurance personnel need to create risk profiles based on the likelihood of certain individuals to file claims.

RStat bridges the gap between the rear and forward-facing views of business operations. It offers the first fully integrated BI and data mining environment for developing predictive models and distributing scoring applications so operational users can make decisions with confidence instead of relying on their instincts.


Top of page

x
Benefits of a Fully Integrated Environment

Data mining is the technique of identifying patterns and relationships within large databases through the use of advanced statistical methods. It extracts historical data and then applies statistical techniques to build a model. Traditionally, highly trained analysts and statisticians built these models. But unless their results are widely deployed, they end up as isolated research products, doing little good for the business.

A scoring application deploys analytic models for repeated use on new data sets by non-technical users to support decision-making. For example, a marketing analyst would use a scoring application to score new mailing lists in order to screen for the best possible respondents. In simple terms, the scoring application labels a prospect as either good or bad.

Statisticians spend much of their time extracting and querying data. But, by working in the same BI environment, developers can create queries that statisticians reuse to create models. The statisticians can compile their models as standard WebFOCUS functions that BI developers turn into WebFOCUS scoring applications, deployable on any platform. There is no need to work with multiple tools or pay for extra licenses. By unifying BI and data mining environments, RStat reduces licensing costs by consolidating software tools. This has the corollary effect of simplifying maintenance and making optimum use of IT resources.


WebFOCUS